ns-visual

Original：🇨🇳 Chinese

Translated

Novel Visual Prompt Skill. Used to generate reusable Chinese or English AI drawing prompts for characters, scenes, covers, chapter illustrations, storyboards, props, maps, and atmosphere diagrams; activated when users request illustration prompts, character standees, cover images, scene images, visual settings, consistent art styles, or extracting scenes from the text.

4installs

Sourcehonestman9527/novel-studio

Added on2026-04-29

NPX Install

npx skill4agent add honestman9527/novel-studio ns-visual

SKILL.md Content (Chinese)

View Translation Comparison →

NS Visual

By default, only generate prompts; invoke image generation capabilities only when the user explicitly requests to generate images.

Read

```
art.yaml
```
```
memory.yaml
```
Current chapter
```
## Main Text
```
Read prompt-patterns.md when specifying model formats

Output

Provide both Chinese and English prompts by default.
Follow the target format when specifying Midjourney, SDXL, or natural language image models.
Split into stable elements and variable elements when reusing characters/scenes.
Write complete prompts, negative prompts, parameters, and iteration records to
```
visuals/*.md
```
.
Summarize visual stable elements, prompt files, and media file paths into
```
art.yaml
```
.